Prestressed concrete beams with stirrups subjected to torsion, bending and shear

Reported in this thesis are test results of 37 eccentrically prestressed beams with stirrups. Single variable parameters were investigated including the prestressing force, the prestressing steel area, the concrete strength, the aspect ratio h/b and the stirrups size and spacing. Interaction of bending, torsion and shear was also investigated by testing a series of beams subjected to varying bending/torsional moment ratios. For the torsional strength an empirical expression of linear format is proposed and can be rearranged in a non-dimensional interaction form: T/To+V/Vo+M/Mo+Ps/Po+Fs/Fo=Pc2/Fsp. This formula which is based on an average experimental steel stress lower than the yield point is compared with 243 prestressed beams containing ' stirrups, including the author's test beams, and good agreement is obtained. For the theoretical analysis of the problem of torsion combined with bending and shear in concrete beams with stirrups, the method of torque-friction is proposed and developed using an average steel stress. A general linear interaction equation for combined torsion with bending and/or shear is proposed in the following format: (fi) T/Tu=1 where (fi) is a combined loading factor to modify the pure ultimate strength for differing cases of torsion with bending and/or shear. From the analysis of 282 reinforced and prestressed concrete beams containing stirrups, including the present investigation, good agreement is obtained between the method and the test results. It is concluded that the proposed method provides a rational and simple basis for predicting the ultimate torisional strength and may also be developed for design purposes.

A hierarchical latent variable model for data visualization

Visualization has proven to be a powerful and widely-applicable tool the analysis and interpretation of data. Most visualization algorithms aim to find a projection from the data space down to a two-dimensional visualization space. However, for complex data sets living in a high-dimensional space it is unlikely that a single two-dimensional projection can reveal all of the interesting structure. We therefore introduce a hierarchical visualization algorithm which allows the complete data set to be visualized at the top level, with clusters and sub-clusters of data points visualized at deeper levels. The algorithm is based on a hierarchical mixture of latent variable models, whose parameters are estimated using the expectation-maximization algorithm. We demonstrate the principle of the approach first on a toy data set, and then apply the algorithm to the visualization of a synthetic data set in 12 dimensions obtained from a simulation of multi-phase flows in oil pipelines and to data in 36 dimensions derived from satellite images.